Results 1 to 16 of 16

Thread: Measuring Performance

  1. #1

    Measuring Performance

    What is typically used to measure system performance of MD5CRK, and how do I find out what mine is? Thanks.

  2. #2
    Sieve it, baby!
    Join Date
    Nov 2002
    Location
    Potsdam, Germany
    Posts
    959

    Re: Measuring Performance

    Originally posted by Jeff321
    What is typically used to measure system performance of MD5CRK, and how do I find out what mine is? Thanks.
    Performance is measured in MegaMD5/sec, which is the amount of million MD5 hashes you compute each second. It can be found in the line "Current Rate:" of the client (assuming you use the windows one - I'm unsure whether other clients have GUIs).

  3. #3
    Ok, thanks. Wasn't sure with all those numbers.
    Mine's 11.86... hope that's good (I have a 3.07ghz P4 )

  4. #4

    <-- Newb

    Mind explaining what this means also...


    <!-- Notice to all: modifying this file could be tragic!!!
    Pretty to see, nice to hold
    But if you change this, consider yourself told -->
    <md5crackTestbed>
    <fullname>md5crk_1.0.6_win32</fullname>
    <versionNumber>5</versionNumber>
    <buildDate>Apr 20 2004 14:17:01</buildDate>
    <md5coreX86ANSIC>
    <rate>4.773034</rate>
    <time>3515</time>
    <xforms>16777216</xforms>
    <cycles>0</cycles>
    <passed>yes</passed>
    </md5coreX86ANSIC>
    <md5coreX86ALU>
    <rate>5.450687</rate>
    <time>3078</time>
    <xforms>16777216</xforms>
    <cycles>615</cycles>
    <passed>yes</passed>
    </md5coreX86ALU>
    <md5coreX86MMX>
    <rate>4.493095</rate>
    <time>7468</time>
    <xforms>33554432</xforms>
    <cycles>1270</cycles>
    <passed>yes</passed>
    </md5coreX86MMX>
    <md5coreX86MMXALU>
    <rate>7.158533</rate>
    <time>7031</time>
    <xforms>50331648</xforms>
    <cycles>1291</cycles>
    <passed>yes</passed>
    </md5coreX86MMXALU>
    <md5coreX86MMXALU2>
    <rate>9.004275</rate>
    <time>7453</time>
    <xforms>67108864</xforms>
    <cycles>1354</cycles>
    <passed>yes</passed>
    </md5coreX86MMXALU2>
    <md5coreX86SSE2>
    <rate>9.987924</rate>
    <time>6719</time>
    <xforms>67108864</xforms>
    <cycles>1184</cycles>
    <passed>yes</passed>
    </md5coreX86SSE2>
    <md5coreX86SSE2ALU2>
    <rate>13.649260</rate>
    <time>7375</time>
    <xforms>100663296</xforms>
    <cycles>1531</cycles>
    <passed>yes</passed>
    </md5coreX86SSE2ALU2>
    <md5coreX86SSE2ALU4>
    <rate>10.213662</rate>
    <time>13141</time>
    <xforms>134217728</xforms>
    <cycles>2266</cycles>
    <passed>yes</passed>
    </md5coreX86SSE2ALU4>
    <preferedCore>X86SSE2ALU2</preferedCore>
    </md5crackTestbed>

  5. #5
    Sieve it, baby!
    Join Date
    Nov 2002
    Location
    Potsdam, Germany
    Posts
    959
    That is the result of the speed test that is performed when you first start the client.
    According to these measurements, it can select the fastest core for that system.

    P.S.:
    My P4c 3 GHz (PSB800; HT enabled) does 13 MegaMD5/sec with one instance and a combined 15 with 2 instances of the client.
    The performance of several systems is posted here . Basically, P4s perform quite well, G5s beat everything by a wide margin...

  6. #6
    Cool... I should try running 2 then, I have HT too. Is there anything special I need to do or just start it twice?

  7. #7
    Originally posted by Jeff321
    Cool... I should try running 2 then, I have HT too. Is there anything special I need to do or just start it twice?
    You should run two in seperate directories (dont' copy the whole directory, only the md5crk application. new md5_user/send/save files will be created)

  8. #8
    is there a core's also optimize for the athlon (xp's)

    Don't see any use of 3dnow.

    And is sse2 used with mmx or then only sse2.
    visit us at TEAM Grutte Pier [Wa Oars]

    Lekker frysk bierke BEER !!!!
    kin wol sponsor fan us team weze !

  9. #9
    Originally posted by [GP]GP500
    is there a core's also optimize for the athlon (xp's)

    Don't see any use of 3dnow.

    And is sse2 used with mmx or then only sse2.
    sse2 uses the same FPU ports as MMX so they can't be used together.

    3dnow - no we dont' have any 3dnow. Want to write it?

  10. #10
    Originally posted by jlcooke
    sse2 uses the same FPU ports as MMX so they can't be used together.

    3dnow - no we dont' have any 3dnow. Want to write it?
    If i could i would.

    And will there be an 64-bit version for the athlon64 and the upcoming xeon's with with amd64.


    Ps:
    Via eshter core
    "The new C5J Esther core provides world-class security for e-commerce transactions by accelerating RSA encryption and Secure Hashing (SHA-1 and SHA-256), together with support for execution (NX) protection for countering email worms/viruses. Other performance enhancing features include a new faster Front Side Bus (FSB) of up to 800MHz, SSE2/SSE3 multimedia instructions, and a larger L2 cache"

    Would this be usefull for projects like this or just bull
    They or suporting it with special software.

    Maybe an supercracker out of an via C3
    visit us at TEAM Grutte Pier [Wa Oars]

    Lekker frysk bierke BEER !!!!
    kin wol sponsor fan us team weze !

  11. #11
    Originally posted by [GP]GP500
    If i could i would.

    And will there be an 64-bit version for the athlon64 and the upcoming xeon's with with amd64.


    Ps:
    Via eshter core
    "The new C5J Esther core provides world-class security for e-commerce transactions by accelerating RSA encryption and Secure Hashing (SHA-1 and SHA-256), together with support for execution (NX) protection for countering email worms/viruses. Other performance enhancing features include a new faster Front Side Bus (FSB) of up to 800MHz, SSE2/SSE3 multimedia instructions, and a larger L2 cache"

    Would this be usefull for projects like this or just bull
    They or suporting it with special software.

    Maybe an supercracker out of an via C3
    It would have to be a C5J (and not a C5P). And this assumes that the entire SHA-1/SHA-256 compression function is not implemented. There is a good chance that it's just the iterator. I'm looking into this now.

  12. #12
    what about athlon omtimized core's then athlon, xp, 64.

    You seem forget the first part questions constantley
    visit us at TEAM Grutte Pier [Wa Oars]

    Lekker frysk bierke BEER !!!!
    kin wol sponsor fan us team weze !

  13. #13
    Originally posted by [GP]GP500
    what about athlon omtimized core's then athlon, xp, 64.

    You seem forget the first part questions constantley
    Being bust does that.

    I replied - 3dnow needs someone to write it. I cannot - too busy.

    As for AMD64 - same thing. Too busy. find someone and point them at the CVS co-ordinates in http://www.md5crk.com/?sec=howhelp

  14. #14
    Originally posted by jlcooke
    Being bust does that.

    I replied - 3dnow needs someone to write it. I cannot - too busy.

    As for AMD64 - same thing. Too busy. find someone and point them at the CVS co-ordinates in http://www.md5crk.com/?sec=howhelp
    Understand it, just like too get the most out of it.
    And it so silent here .
    visit us at TEAM Grutte Pier [Wa Oars]

    Lekker frysk bierke BEER !!!!
    kin wol sponsor fan us team weze !

  15. #15
    Originally posted by [GP]GP500
    And it so silent here .
    In the next week or two it should get crowded I'm sure.


  16. #16
    Originally posted by jlcooke
    In the next week or two it should get crowded I'm sure.
    Interesting... Very interesting...

    Jeff.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •